A gated piecewise CNN with entity-aware enhancement for distantly supervised relation extraction. Issue 6 (November 2020)

• Record Type: Journal Article
• Authors: Wen, Haixu; Zhu, Xinhua; Zhang, Lanfang; Li, Fei
• Is Part Of: Information processing & management. Volume 57:Issue 6(2020:Nov.)
• Publication Date: 2020-11
• Subjects: Distant supervision -- Relation extraction -- PCNN -- Multi-head self-attention -- Gate mechanism; Information storage and retrieval systems -- Periodicals; Information science -- Periodicals; Systèmes d'information -- Périodiques; Sciences de l'information -- Périodiques; Information science; Information storage and retrieval systems; Periodicals; 658.4038
• DOI: 10.1016/j.ipm.2020.102373
• Languages: English
• ISSNs: 0306-4573

Abstract

Highlights: We propose an entity-aware enhanced word representation with rich context, which enables the downstream modules to learn robust semantic features. We combine the global gate structure and PCNN to better capture the global and local features of the sentence. We introduce a gate mechanism after the max-pooling layer of PCNN model, which assigns different weights to the three segments and highlights the effect of crucial segments. Our model is evaluated on the widely used benchmark dataset and outperforms most of the stateof-the-art methods. Abstract: The piecewise convolutional neural network (PCNN) is an important method for distant supervision relation extraction. However, the existing methods based on the PCNN still have the following shortcomings: these methods lack the consideration of the impacts of entity pairs and the sentence context on word encoding and do not distinguish the different contributions of the three segments in PCNN to relation classification. To solve these problems, we propose a novel gated piecewise CNN with entity-aware enhancement for distantly supervised relation extraction. First, we use a multi-head self-attention mechanism to combine the word embedding with the head/tail entity embedding and relative position embedding to generate an entity-aware enhanced word representation, which is capable of capturing the semantic dependency between each word and entity pair. Then we introduce a global gate to combine each entity-aware enhanced word representation with their average in the input sentence to form the final word representation of the PCNN input. Moreover, to determine the key segments where the most important information for relation classification appears, we design another gate mechanism to assign a different weight to each sentence segment to highlight the effects of key segments on the PCNN. Experiments on New York Times dataset demonstrate that our model significantly outperforms most of the state-of-the-art models.